Too Many Surveys, Too Little Passion?

Gap's heavy use of research has sent designers fleeing and may be hurting sales

When Paul S. Pressler began his effort to turn around Gap Inc. (GPS ) in late 2002, he jumped on one task right away: getting to know the company's customers better. The former Walt Disney Co. (DIS ) theme-park executive quickly put together a "consumer insights" staff charged with developing focus groups, surveys, and other market research. Analysts applauded the moves, and Pressler's approach seemed to work well. After faltering for three years under predecessor Millard S. "Mickey" Drexler -- a merchant famous for relying on his gut fashion instinct -- same-store sales grew for six straight quarters under Pressler.

But if a little research discipline was needed, Pressler may have overdone it. Over the past year sales at the San Francisco operator of Gap, Banana Republic, and Old Navy have been sliding. In the fiscal quarter ended on Apr. 30, sales fell 4%. Analysts expect them to drop 2% more for the quarter ended on July 30. The reason, according to eight former employees and two analysts: Pressler has shifted too far toward research and away from the instinct and emotion favored by many successful clothing merchandisers.

Gap disputes this notion. Market research "guides but does not dictate" choices, says a spokeswoman. The company also says a Gap-brand store redesign, based in part on consumer comments and launched three months ago in the Denver area, is getting a good response. But the shift toward research is roiling the creative staff. In the past 10 months at least nine designers and merchandisers have left -- saying Gap has lost its creative spark. The emphasis on research, says one, has "distracted from the ability to create and take risks."

Every retailer misreads the market some of the time. It's no surprise, then, that all apparel makers use market research to varying degrees. But Gap has gone further, even hiring economists and statisticians to study everything from consumer trends to store traffic patterns.

Research has inherent shortcomings in the clothing business. Because fashion changes every few months, it can get stale fast. And since consumers can't predict how future trends will influence them, they may like a dress in a fashion mag -- yet buy a different one. "It isn't always true [that] customers know what they want," says Susan Rolontz, executive vice-president of research firm Tobé.


Gap insists it's no slave to research. Sheryl Clark, Old Navy's merchandising chief, says it "doesn't lead my decisions," but it influences them. And Pressler has cited the new test stores as evidence that consumer insights help. Based in part on conversations with customers, Gap designed the stores with a woodsy ambience and separate entrances to emphasize the different feel of the men's and women's departments. While Gap won't say whether sales at the new stores are growing faster than at older ones, Pressler told analysts in May that "fitting-room usage is up considerably, and customers are spending more time in the stores on each visit."

So could an overhaul of the company's 2,759 stores turn things around? Not without better merchandise, say the former staffers. And that means less emphasis on research, which, they say, has led to some missteps. Last year, say three former employees, Old Navy designers proposed a line of lightweight T-shirts, only to be told by researchers that some consumers associate lightweight fabrics with poor quality. So Old Navy ordered only a few, which ended up selling well. Now Old Navy stocks plenty of lightweight T-shirts. Gap says it was simply responding to an industrywide trend.

Some analysts agree Gap has taken data- gathering too far. "They should let the merchants work off their instincts," says Wells Fargo Securities LLC (WFC ) analyst Mark Montagna. Pressler and his researchers may have to accept what merchandisers learned long ago: As often as not, it takes creative talent to figure out what folks will buy.

By Louise Lee in San Mateo, Calif.

    Before it's here, it's on the Bloomberg Terminal.